The development of clinical autoimmune disease comes after months or years of subclinical autoantibody accumulation. As one example, antinuclear antibodies appear 3.7 or more years before systemic lupus erythematosus (SLE) begins, antibodies to double-stranded DNA at 2.2 years before disease, and anti-Sm and anti-nuclear RNP antibodies in the year SLE is diagnosed. The long time frame has made it difficult to study early changes in gene expression leading up to the development of clinical disease. However, we propose to take advantage of a dramatic compression of this time course that can occur in patients treated with TNFalpha inhibitors, who develop a positive ANA in up to 71% of cases, and positive anti-dsDNA antibodies in up to 17% of cases. These antibodies frequently develop within weeks to months, making a prospective investigation possible. Actual clinical cases of lupus have been unusual but do occur. Our hypothesis is that this rapid induction of autoantibodies provides a unique opportunity to prospectively identify the critical gene expression changes that characterize autoimmune disease. In this proposal, we will use our large populations of patients with rheumatoid arthritis and spondyloarthopathies to enroll those with a negative ANA who are about to begin a TNFa inhibitor. We will perform FACS analysis to characterize the PBMC phenotype in detail, followed by microarray analysis of PBMC subsets (B cells, monocytes/macrophages, CD4+ T cells, CD8+ T cells). Comparisons of baseline with those after the ANA and later the anti-dsDNA liters become positive will establish a gene expression profile of the earliest features of systemic autoimmunity, even in the absence of clinical SLE. Our emphasis on using four PBMC subsets in microarrays reflects our observations that alterations in gene expression are more dramatically demonstrable when such pure populations of cells are studied. The further characterization of candidate genes will use RTPCR, Northern blotting, and Western blotting to validate our findings. The gene expression signatures of autoantibody induction that emerges from this study will characterize the earliest stages of autoimmune disease induction.